This Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) provides $149,343 to the University of Nebraska for the development of a quantitative verification tool for learning-enabled cyber-physical systems (LE-CPS). The key products to be delivered include:
A new approach for qualitative and quantitative verification of temporal properties of LE-CPS at the system level, including exact and approximate verification schemes that can provide precise probability of safety properties being satisfied.
The development of ProbStar Temporal Logic (PSTLs) to enable quantitative verification of temporal properties of LE-CPS, along with associated reachability algorithms.
A user-friendly interface and Robotic Operating System (ROS) integration to facilitate the adoption of the new verification techniques into real robotic applications.
The project aims to address the critical industry need for rigorous verification methods to ensure the reliability and safety of ML-based cyber-physical systems, with a focus on applications like autonomous vehicles. The award period runs from October 1, 2024 to September 30, 2026.